Developing probabilistic approach for asphaltic overlay design by considering variability of input parameters

Abstract

Abstract The asphaltic overlay design process considers various input parameters. These parameters are probabilistic in nature, but the current design process adopts deterministic values. This study proposes a comprehensive probabilistic approach for asphaltic overlay design to accommodate the variations in pavement layer thicknesses, pavement layer moduli, vehicle damage factor (VDF), lane distribution factor (LDF) and vehicle growth rate (r). An analytical method, first order second moment (FOSM), is used in developing the proposed design approach. A design example based on real world falling weight deflectometer (FWD) data of a 30 km long highway stretch is worked out to illustrate the proposed design approach. The results indicate that the current deterministic approach estimates an overlay thickness with approximately 95% reliability. Detail analysis of results from the proposed probabilistic approach denotes that choosing marginally lower design reliability could significantly reduce the overlay thickness. A sensitivity analysis of pavement layer moduli indicated that the asphaltic overlay thickness is sensitive to resilient modulus of base layer for fatigue and subgrade layer for rutting. The probability based design approach can accommodate variability of field material properties and construction practices in the asphaltic overlay design process.